Pokémon Challenge

Submitted by: Student

From school: A High School

Introduction

In this notebook, I will be comparing the strongest 15 Pokémon in the categories: "attack", "sp_attack", "defense", "sp_defense", "hp", and "speed" with each other, along displaying these attributes against regular Pokémon compared to legendary Pokémon! As someone new to CS and data science, this isn't the most elaborate hackathon submission. Nonetheless, enjoy my findings!

# Import libraries and data
import pandas as pd
import plotly.express as px
from IPython.display import YouTubeVideo
pokemon = pd.read_csv('https://raw.githubusercontent.com/callysto/data-files/main/hackathon/pokemon.csv')
pokemon

	abilities	against_bug	against_dark	against_dragon	against_electric	against_fairy	against_fight	against_fire	against_flying	against_ghost	...	percentage_male	pokedex_number	sp_attack	sp_defense	speed	type1	type2	weight_kg	generation	is_legendary
0	['Overgrow', 'Chlorophyll']	1.00	1.0	1.0	0.5	0.5	0.5	2.0	2.0	1.0	...	88.1	1	65	65	45	grass	poison	6.9	1	0
1	['Overgrow', 'Chlorophyll']	1.00	1.0	1.0	0.5	0.5	0.5	2.0	2.0	1.0	...	88.1	2	80	80	60	grass	poison	13.0	1	0
2	['Overgrow', 'Chlorophyll']	1.00	1.0	1.0	0.5	0.5	0.5	2.0	2.0	1.0	...	88.1	3	122	120	80	grass	poison	100.0	1	0
3	['Blaze', 'Solar Power']	0.50	1.0	1.0	1.0	0.5	1.0	0.5	1.0	1.0	...	88.1	4	60	50	65	fire	NaN	8.5	1	0
4	['Blaze', 'Solar Power']	0.50	1.0	1.0	1.0	0.5	1.0	0.5	1.0	1.0	...	88.1	5	80	65	80	fire	NaN	19.0	1	0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
796	['Beast Boost']	0.25	1.0	0.5	2.0	0.5	1.0	2.0	0.5	1.0	...	NaN	797	107	101	61	steel	flying	999.9	7	1
797	['Beast Boost']	1.00	1.0	0.5	0.5	0.5	2.0	4.0	1.0	1.0	...	NaN	798	59	31	109	grass	steel	0.1	7	1
798	['Beast Boost']	2.00	0.5	2.0	0.5	4.0	2.0	0.5	1.0	0.5	...	NaN	799	97	53	43	dark	dragon	888.0	7	1
799	['Prism Armor']	2.00	2.0	1.0	1.0	1.0	0.5	1.0	1.0	2.0	...	NaN	800	127	89	79	psychic	NaN	230.0	7	1
800	['Soul-Heart']	0.25	0.5	0.0	1.0	0.5	1.0	2.0	0.5	1.0	...	NaN	801	130	115	65	steel	fairy	80.5	7	1

801 rows × 41 columns

Column Descriptions:

• name: The English name of the Pokémon
• japanese_name: The Original Japanese name of the Pokémon
• pokedex_number: The entry number of the Pokémon in the National Pokédex
• percentage_male: The percentage of the species that are male. Blank if the Pokémon is genderless.
• type1: The Primary Type of the Pokémon
• type2: The Secondary Type of the Pokémon
• classfication: The Classification of the Pokémon as described by the Sun and Moon Pokédex
• height_m: Height of the Pokémon in metres
• weight_kg: The Weight of the Pokémon in kilograms
• capture_rate: Capture Rate of the Pokémon
• base_egg_steps: The number of steps required to hatch an egg of the Pokémon
• abilities: A stringified list of abilities that the Pokémon is capable of having
• experience_growth: The Experience Growth of the Pokémon
• base_happiness: Base Happiness of the Pokémon
• against_?: Eighteen features that denote the amount of damage taken against an attack of a particular type
• hp: The Base HP of the Pokémon
• attack: The Base Attack of the Pokémon
• defense: The Base Defense of the Pokémon
• sp_attack: The Base Special Attack of the Pokémon
• sp_defense: The Base Special Defense of the Pokémon
• speed: The Base Speed of the Pokémon
• generation: The numbered generation which the Pokémon was first introduced
• is_legendary: Denotes if the Pokémon is legendary.

Comparison 1: Fighting - Attacks

Pokémon are always used in fighting tournaments. Let's create a chart that displays the top 15 Pokémon with the strongest Base Attack, accompanied by their Base Special Attack!

pokemon_by_attack = pokemon[["name", "attack", "sp_attack"]]
pokemon_by_attack = pokemon_by_attack.sort_values(by=["attack"], ascending=False)
pokemon_by_attack.head(15)

	name	attack	sp_attack
213	Heracross	185	40
797	Kartana	181	59
382	Groudon	180	150
383	Rayquaza	180	180
444	Garchomp	170	120
408	Rampardos	165	65
353	Banette	165	93
474	Gallade	165	65
247	Tyranitar	164	95
288	Slaking	160	95
718	Diancie	160	160
485	Regigigas	160	80
719	Hoopa	160	170
256	Blaziken	160	130
126	Pinsir	155	65

Visualizing the Data

Let's visualize the chart by turning it into a bar graph!

px.bar(pokemon_by_attack.head(15), x="name", y=["attack", "sp_attack"], barmode="group", title="Top 15 Strongest Pokémon by Attack, Accompanied by Their SP Attack")

Interestingly, Rayquaza has the most powerful Base Special Attack despite only having the 4th strongest Base Attack. Furthermore, Heracross, having the strongest Base Attack out of all Pokémon, had the weakest Base Special Attack within the top 15 Pokémon on the graph.

Now let's graph all Pokémon (within the dataset) based on their Base Attack and Base SP Attack. Furthermore, let's identify which Pokémon are legendary!

px.scatter(pokemon, x='attack', y='sp_attack', color="is_legendary", hover_data=['name'], title='Pokémon Attack vs SP Attack')

Comparison 2: Fighting - Defense

A Pokémon can't participate in a true tournament without some form of defense. Let's create a new chart that displays the top 15 Pokémon with the strongest Base Defense, accompanied by their Base Special Defense!

pokemon_by_defense = pokemon[["name", "defense", "sp_defense"]]
pokemon_by_defense = pokemon_by_defense.sort_values(by=["defense"], ascending=False)
pokemon_by_defense.head(15)

	name	defense	sp_defense
305	Aggron	230	80
212	Shuckle	230	230
207	Steelix	230	95
376	Regirock	200	100
712	Avalugg	184	46
90	Cloyster	180	45
79	Slowbro	180	80
410	Bastiodon	168	138
94	Onix	160	45
382	Groudon	160	90
747	Toxapex	152	142
679	Doublade	150	49
378	Registeel	150	150
247	Tyranitar	150	120
375	Metagross	150	110

Visualizing the Data 2

Repeating what we did with the strongest Pokémon attacks, let's do the same with the strongest Pokémon defense and comparing these variables on all Pokémon with a scatter graph!

px.bar(pokemon_by_defense.head(15), x="name", y=["defense", "sp_defense"], barmode="group", title="Top 15 Strongest Pokémon by Defense, Accompanied by Their SP Defense")

px.scatter(pokemon, x='defense', y='sp_defense', color="is_legendary", hover_data=['name'], title='Pokémon Defense vs SP Defense')

Unlike the first scatter graph, where the difference between legendary Pokémon and regular Pokémon was wider in regards to Base Attacks and Base Special Attacks, the gap is a lot closer when it comes to Base Defense and Base Special Defense

Comparison 3: Fighting - HP & Speed

Repeating what we did in the last 2 comparisons, let's finally see who the best Pokémon are when it comes to HP and Speed!

pokemon_by_remainder = pokemon[["name", "hp", "speed"]]
pokemon_by_remainder = pokemon_by_remainder.sort_values(by=["hp"], ascending=False)
pokemon_by_remainder.head(15)

	name	hp	speed
241	Blissey	255	55
112	Chansey	250	50
798	Guzzlord	223	43
717	Zygarde	216	85
201	Wobbuffet	190	33
320	Wailord	170	60
593	Alomomola	165	65
142	Snorlax	160	30
486	Giratina	150	90
425	Drifblim	150	80
288	Slaking	150	100
296	Hariyama	144	50
39	Wigglytuff	140	45
790	Solgaleo	137	97
791	Lunala	137	97

Visualizing the Data 3

px.bar(pokemon_by_remainder.head(15), x="name", y=["hp", "speed"], barmode="group", title="Top 15 Strongest Pokémon by HP, Accompanied by Their Speed")

Not surprisingly, there seems to be somewhat of a correlation between a Pokémons HP, and their Speed. This may be because a Pokémon with more HP would be more heavy and typically slower than other types of Pokémon. Such a hypothesis is supported with Pokémon like Blissey, Chansey, and Guzzlord having high HP but low Speed.

px.scatter(pokemon, x='hp', y='speed', color="is_legendary", hover_data=['name'], title='Pokémon HP vs Speed')

Hackathon Reflections

• What is something you learned through this process?
• I learned a significant amount of Python code while participating in this hackathon! As a student who has never had any experience with data science and coding in general, this hackathon has given me a new opportunity to learn how data science works, along with helping me improve my Python coding skills (one way or another).
• How well did your group work together? Why do you think that is?
• As I worked by myself, this question doesn't apply to me.
• What were some of the hardest parts?
• Learning how to properly graph my charts and graphs was always an issue for me. For example, figuring out how to create the duel graph comparing Attack and SP Attack.
• What are you proud of? What would you like to show others?
• I feel extremely proud that I even participated in the hackathon in the first place. I'm sure my notebook isn't the most detailed considering a few of my friends are wonders ahead of me with coding experience. That, along with having to google how to do things with pandas. However, I'm proud of the Pokémon notebooks I've created on my own.
• Are you curious about anything else related to this? Did anything surprise you?
• With this hackathon, I'm curious about Jupyter and Python in the realm of data science. While it wasn't a surprise, I still found it slightly challenging (in a fun way) to create this. I wonder how hard data science can really get in the new world.
• How can you apply your learning to future activities?
• In can apply my learning in my future Computer Science (AP) classes at school, along with creating personal projects to further grow my coding expertise!